The central hypothesis is that effective cancer gene therapy can be achieved using a highly toxic transgene that is expressed controllably from a heat-inducible promoter. To test this, cytolethal distending toxin B (CdtB), the cytotoxic component of a tripartite bacterial proteotoxin, was placed into an adenovirus vector under control of a modified human HSP70B heat shock promoter that is extremely silent until heal shocked at 41.0 degrees C or higher. Aim 1 is to optimize using this heat-activatable CdtB gene therapy to establish local tumor control in a rabbit VX2 brain tumor model and to protract animal survival. Aim two is to deliver the adenovirus vectors with tissue permeabilizing agents that increase the interstitial space to promote vector diffusion throughout solid tumors to ascertain if this increases treatment efficacy. It is postulated that using the permeabilizers will increase the fraction of tumor cells that get infected to express CdtB to control tumors more efficiently and permit larger tumors to be treated successfully. Aim three is to explicate the mechanisms of CdtB bystander killing. Treatment efficacy is dependent upon bystander killing of cells adjacent to those infected with and expressing the CdtB transgene. Although CdtB bystander killing has been observed in vitro and in vivo, little is known about it. Experiments will determine if bystander killing is mediated by a freely diffusible extracellular signal or if cell-cell contact and gap junction communication are requisite. Assays will also establish if the bystander cells die by apoptosis or other death mechanisms and will identify the signals that initiate bystander killing. Some tumors, e.g., glioblastoma multiforme, pancreatic tumors, etc. are more refractory to conventional radiotherapy, and chemotherapy, making local tumor control difficult to achieve. Additionally, many tumors recur locally with high frequency and surgery is often impossible and/or radiotherapy options are limited because surrounding normal tissues have accrued their tolerance radiation dose. Consequently, complementary therapies that can improve local control for primary and recurrent cancers are needed. The proposed study will test the potential of heat-activated CdtB gene therapy to help satisfy this need.